David Mitchell [Tue, 2 Mar 2010 16:26:36 +0000]
fetching from e.g. github requires a repo name
Craig A. Berry [Tue, 2 Mar 2010 04:01:48 +0000]
Skip Pod::Parser's find.t on VMS for now.
Pending integration of https://rt.cpan.org/Ticket/Display.html?id=55121
Craig A. Berry [Tue, 2 Mar 2010 01:55:08 +0000]
Don't scrub DCL$PATH unless it's there.
It's optional and this was causing test failures when it didn't exist.
Craig A. Berry [Tue, 2 Mar 2010 00:56:45 +0000]
Cannot portably split on $Config{path_sep} -- use quotemeta($Config{path_sep}).
Broken in
fc5e5837c991d3d3224259ff5c1d728d4e0636e2.
On VMS we were getting:
$ perl -"MConfig" -e "print join('+',split $Config{path_sep}, 'foo|bar|baz');"
f+o+o+|+b+a+r+|+b+a+z
which is a *lot* more pieces than we want. What was intended was:
$ perl -"MConfig" -e "print join('+',split quotemeta($Config{path_sep}), 'foo|bar|baz');"
foo+bar+baz
No version bump as this is test infrastructure and 1.55 has not been
released outside of development releases of blead.
Karl Williamson [Sun, 28 Feb 2010 15:48:33 +0000]
Note this doesn't work: $re = "\\N{...}"; /$re/
Karl Williamson [Fri, 26 Feb 2010 03:45:11 +0000]
Document some re bugs in perlre.pod
Karl Williamson [Fri, 26 Feb 2010 03:44:04 +0000]
Document 5.8 regression #72998 in delta
Karl Williamson [Fri, 26 Feb 2010 02:38:15 +0000]
Document parsing changes for [perl #56444] patch
Karl Williamson [Fri, 26 Feb 2010 02:17:57 +0000]
Revise wording about /x caveats
Karl Williamson [Fri, 26 Feb 2010 02:16:56 +0000]
Mention \N{U+...} documentation in delta
Karl Williamson [Fri, 26 Feb 2010 02:14:30 +0000]
Add statement about UTF-8 and \N{} to delta
Karl Williamson [Fri, 26 Feb 2010 01:48:50 +0000]
Document Unicode case-insensitive [] range bug
Karl Williamson [Fri, 26 Feb 2010 01:39:29 +0000]
Quote a code example
Karl Williamson [Thu, 25 Feb 2010 23:28:50 +0000]
Add deprecation of non-name \N{foo} to delta pod
Karl Williamson [Thu, 25 Feb 2010 22:03:47 +0000]
Revise notes on /x modifier
Karl Williamson [Thu, 25 Feb 2010 21:43:48 +0000]
Mark \N meaning [^\n] as experimental
Karl Williamson [Thu, 25 Feb 2010 20:41:10 +0000]
Mention there are places /x modifier is ineffective
Karl Williamson [Thu, 25 Feb 2010 19:49:31 +0000]
Clarify perlrebackslash.pod
Karl Williamson [Thu, 25 Feb 2010 19:48:12 +0000]
Note existence (and warn against using) \N{U+c1.c2...}
Karl Williamson [Thu, 25 Feb 2010 19:36:53 +0000]
Note that \N{U+...} forces character semantics
Karl Williamson [Thu, 25 Feb 2010 19:35:14 +0000]
Document \N{U+...}
Karl Williamson [Thu, 25 Feb 2010 19:31:12 +0000]
Mention \N{U+...} in perlunicode.pod
Karl Williamson [Thu, 25 Feb 2010 19:25:04 +0000]
Update documentation
List known bugs, mention new meaning of \N
Karl Williamson [Wed, 24 Feb 2010 00:33:35 +0000]
Update pods for \N changes
Karl Williamson [Wed, 24 Feb 2010 00:31:48 +0000]
Update charnames documentations for \N changes, bugs
\N has a possible new meaning, and mention bug reports filed against
charnames
Karl Williamson [Tue, 23 Feb 2010 04:43:35 +0000]
Clarify sv.c API wording.
David Mitchell [Sun, 28 Feb 2010 15:13:33 +0000]
fix for [perl #72604] @DB::args and win32 fork
A previous fix for [perl #66108] (
7fa38291524c327a3cb23bfe94979e1537743cac)
stopped cloning PL_dbargs, on the grounds that it was usually filled with
garbage (it contains an un-refcounted copy of @_'s elements; once the
function has returned, these may have been freed or reassigned). However,
the fix instead recreated PL_dbargs as a new empty AV that *wasn't* then
associated with the DB::args glob; so modifications to PL_dbargs weren't
seen via @DB::args.
The fix is to simply set it to null when cloning; pp_caller() will
recreate it again if necessary when it is needed.
David Mitchell [Fri, 26 Feb 2010 12:50:21 +0000]
bytes.pm: downgrade DEPRECATED to NOTICE
Rafael Garcia-Suarez [Thu, 25 Feb 2010 07:21:07 +0000]
Typo fix
David Mitchell [Wed, 24 Feb 2010 22:41:09 +0000]
more prose fixups to bytes.pm deprecation warning
(suggested by Dr.Rudd)
Jan Dubois [Tue, 23 Feb 2010 20:48:58 +0000]
Teach bump-perl-version about additional perl511 references in makedef.pl.
The previous version also missed 2 instances in win32/makefile.mk because
they were written as "perl511$(a)" instead of "perl511.o".
Nicholas Clark [Tue, 23 Feb 2010 14:23:31 +0000]
Regression test for
34394ecd - SVs that were only on the tmps stack leaked.
Jan Dubois [Tue, 23 Feb 2010 08:54:46 +0000]
Symbol S_no_symref_sv should be static (local to the compilation unit).
Jan Dubois [Tue, 23 Feb 2010 02:27:13 +0000]
Teach bump-perl-version about perl511.dll, perl511.lib and perl511s.lib
Steve Hay [Mon, 22 Feb 2010 23:24:59 +0000]
Update the release managers' guide: don't push the tag too soon!
Remind release managers not to push a new tag until the indexing of
the new uploads has been successful. A problem with indexing 5.11.5
meant that we nearly ended up with 5.11.6 instead, since the 5.11.5
tag had already been pushed.
David Mitchell [Mon, 22 Feb 2010 17:19:43 +0000]
fix prose from previous lib/bytes.pm doc update
David Mitchell [Mon, 22 Feb 2010 16:44:59 +0000]
Better explain utf8::upgrade/downgrade/encode/decode
Try to make it clear what each does, and the fact that upgrade/downgrade
don't change the logical string value while encode/decode do.
David Mitchell [Mon, 22 Feb 2010 15:25:16 +0000]
deprecate bytes.pm
Abigail [Mon, 22 Feb 2010 14:36:50 +0000]
Bram pointed out the new tests hang if the fix for #72878 isn't in place.
This commits traps the recursion and throws an exception.
Steve Hay [Mon, 22 Feb 2010 12:35:45 +0000]
Another minor update to the release managers' guide
Craig A. Berry [Mon, 22 Feb 2010 04:26:15 +0000]
Don't multiply define Perl_regcurly in core and the re extension.
Broken in
ff3f963aa0f95ea53996b6a3842b824504b57c79.
David Mitchell [Sat, 20 Feb 2010 21:01:11 +0000]
stop skipping ext/Fcntl/t/syslfs.t
Since the module reorganisation and now that tests run cd'ed to
ext/Fcntl, expecting to run ../perl no longer works.
Also, for the benefit of future generations, make the inability
to find perl a failure rather than a skip, with a more relevant error
message.
David Mitchell [Sat, 20 Feb 2010 20:03:54 +0000]
stop cpan/Memoize/t/tie_ndbm.t being skipped
Back in 2006,
0b40f9f8cf631862db10aed9706454c7518fef19 added a skip check
for $ENV{TMP} being writeable. This test failed if $TMP wasn't set, but
also, since then this test script has been updated to write its temp files
to the current dir rather than /tmp, so the TMP check is redundant, and is
now removed.
David Mitchell [Sat, 20 Feb 2010 19:34:05 +0000]
stop op/groups.t skipping on Linux
My linux (with selinux) includes a "context=" field as part of the
output of "id -a". Make the extraction of the "groups=(.+)" entry
of "id -a" more robust in the presence of unpredictable (in name and
ordering) fields, by splitting on fieldname=(.*), then looking for a field
name of 'groups'.
David Mitchell [Sat, 20 Feb 2010 16:02:58 +0000]
Sys-Syslog/t/constants.t couldn't find macros.all
The move from ext/ to cpan/ hadn't been reflected in the pathname
of macros.all. However, since tests are now run from the src dir,
the whole $ENV{PERL_CORE} bit is no longer necessary.
(Note that this line has already been modified from the CPAN version,
so a further change to it doesn't really matter as regards maintaining
sync with CPAN).
Steve Hay [Sun, 21 Feb 2010 20:30:34 +0000]
Update Porting/release_managers_guide.pod and Porting/makerel
In the course of releasing perl-5.11.5 I found a few things that are not
quite correct or no longer relevant:
- Add comments on the problems with doing this on Win32
- There are no longer Roman numerals in README.vms (since
eb505c2afb6b5ac2aaba7935878ae6494d44732a)
- Note that Module::CoreList's Changes and META.yml files need updating too
- Fix Porting/makerel's null device redirection
One bigger problem remains to be resolved: when run in a normal cmd.exe
shell on Win32, the Porting/makerel script currently generates tarballs
containing world-writable files and directories, which the PAUSE indexer
then rejects as a possible security risk.
Steve Hay [Sun, 21 Feb 2010 19:44:23 +0000]
Fix the NAME in two old perlXYZdelta.pod files
Steve Hay [Sun, 21 Feb 2010 19:43:45 +0000]
Create perl5116delta and update other versions
Steve Hay [Sun, 21 Feb 2010 19:41:49 +0000]
Stop pod/buildtoc from changing EOLs on Win32
Abigail [Sun, 21 Feb 2010 16:28:39 +0000]
Merge branch 'tie-scalar' into blead
Steve Hay [Sun, 21 Feb 2010 16:20:46 +0000]
Update future release schedule
Put myself down for another release. Hopefully we can get Porting/makerel
and/or the PAUSE indexer sorted out before then to avoid the trouble
experienced rolling perl-5.11.5 from Win32 :-)
Abigail [Sun, 21 Feb 2010 16:13:27 +0000]
Upped VERSION
Nicholas Clark [Sun, 21 Feb 2010 14:46:27 +0000]
Since change
9c901649534a1360, vendorlib_stem isn't always added to @INC.
Update lib/Config.t to reflect this.
vendorlib_stem used always to be added if defined. Since that change it is only
added if inc_version_list is also defined, which makes "vendor" behaviour
consistent with "site", where sitelib_stem had always only ever been added to
@INC if inc_version_list is defined.
Steve Hay [Sat, 20 Feb 2010 21:08:25 +0000]
Final update to perl5115delta.pod
Steve Hay [Sat, 20 Feb 2010 18:34:56 +0000]
Record Perl 5.11.5 and a new pumpkin keeper in perlhist
Steve Hay [Sat, 20 Feb 2010 18:21:13 +0000]
Fix Module::CoreList's own $VERSION in 5.11.4
Steve Hay [Sat, 20 Feb 2010 18:20:18 +0000]
Update Module::CoreList for Perl 5.11.5
Abigail [Sat, 20 Feb 2010 17:55:06 +0000]
Don't recurse forever if both new() and TIESCALAR() are missing.
This should fix issue #72878. Before calling $pkg -> new in TIESCALAR,
we check whether $pkg -> new is actually the new defined in the Tie::Scalar
package; if true, croak instead of calling it.
Abigail [Sat, 20 Feb 2010 17:54:15 +0000]
Document differences between Tie::Scalar and Tie::StdScalar
Steve Hay [Sat, 20 Feb 2010 16:55:22 +0000]
Missed a version bump in README.vms
Steve Hay [Sat, 20 Feb 2010 16:28:04 +0000]
Bump version to 5.11.5
Steve Hay [Sat, 20 Feb 2010 16:22:41 +0000]
Stop Porting/bump-perl-version from changing EOLs on Win32
Steve Hay [Sat, 20 Feb 2010 15:41:36 +0000]
Almost the final updates to perl5115delta.pod
Steve Hay [Sat, 20 Feb 2010 15:40:52 +0000]
Update Porting/Maintainers.pl with version bumps that have been forgotten
This should also bump version to 0.82, but only 0.80 is actually on CPAN.
Likewise, Pod::Perldoc should say 3.15_02, but only 3.15_01 is on CPAN.
David Mitchell [Sat, 20 Feb 2010 15:41:38 +0000]
unTODO some passing TODO tests in reg_fold.t
As far as I can tell these particular tests were never actually TODO,
but were lumped in with other failing tests for ease of coding the .t file
Marcus Holland-Moritz [Sat, 20 Feb 2010 14:33:05 +0000]
Fix apidoc for HeUTF8()
HeUTF8() onlu takes one argument.
Steve Hay [Sat, 20 Feb 2010 12:00:30 +0000]
Update perl5115delta.pod with new diagnostics
Karl Williamson [Sat, 20 Feb 2010 06:53:36 +0000]
PATCH: deprecation warnings for unreasonable charnames
Prior to now just about anything has been legal for a character name in
\N{...}. This means that legal code was broken by having \N{3,4} for
example mean [^\n]{3,4}. Such code doesn't come from standard
charnames, but from legal custom translators.
This patch deprecates "unreasonable" names. handy.h is changed by the
addition of macros that taken together define the names we deem
reasonable, namely alpha beginning with alphanumerics and some
punctuations as continuations.
toke.c is changed to parse each name and to raise a warning if any
problematic characters are found.
Some tests and diagnostic documentation are also included.
Karl Williamson [Fri, 19 Feb 2010 21:42:16 +0000]
Improve handling of qq(\N{...}); and /x
It is possible to bypass the lexer's parsing of \N. This patch causes
the regex compiler to deal with that better. The compiler no longer
assumes that the lexer parsed the \N. It generates an error message if
the \N isn't in a form it is expecting, and invalid hexadecimal digits
are now fatal errors, with the position of the error more clearly
marked.
The diagnostic pod has been updated to reflect the new error messages,
with some slight clarifications to the previous ones as well.
Marcus Holland-Moritz [Sat, 20 Feb 2010 10:13:00 +0000]
Add some missing dVAR's
Commits
c3acb9e0760135dfd888c0ee1b415777d784aabc, 867fa1e2da145229b4db2c6e8d5b51700c15f114
and
f0e67a1d29102aa9905aecf2b0f98449697d5af3 added or changed functions that now require a
dVAR declaration to compile with -DPERL_GLOBAL_STRUCT.
Steve Hay [Sat, 20 Feb 2010 00:49:07 +0000]
Regen headers after previous patches
Karl Williamson [Sat, 20 Feb 2010 00:12:20 +0000]
Add X flag to embed.fnc entries.
It has been pointed out that the E flag should be accompanied by the X
flag, and is not for these recently changed entries.
Karl Williamson [Fri, 19 Feb 2010 22:26:24 +0000]
Move some fncs from Public to extensions visible
Run
make regen embed.fnc
on this patch.
Several functions are in the public API, but don't need to be. They are
helpers for the regex engine, and thus need to be visible to extensions
for 'use re "debug"'.
Steve Hay [Sat, 20 Feb 2010 00:29:55 +0000]
More updates to perl5115delta.pod
Rafael Garcia-Suarez [Fri, 19 Feb 2010 09:53:24 +0000]
Avoid returning an undefined SV*
Rafael Garcia-Suarez [Fri, 19 Feb 2010 09:11:43 +0000]
Regen headers after previous patches
Karl Williamson [Thu, 18 Feb 2010 22:06:51 +0000]
Make a missing right brace on \N{ fatal
It was decided that this should be a fatal error instead of a warning.
Also some comments were updated..
Karl Williamson [Thu, 18 Feb 2010 20:41:09 +0000]
PATCH: [perl #56444] delayed interpolation of \N{...}
make regen embed.fnc
needs to be run on this patch.
This patch fixes Bugs #56444 and #62056.
Hopefully we have finally gotten this right. The parser used to handle
all the escaped constants, expanding \x2e to its single byte equivalent.
The problem is that for regexp patterns, this is a '.', which is a
metacharacter and has special meaning that \x2e does not. So things
were changed so that the parser didn't expand things in patterns. But
this causes problems for \N{NAME}, when the pattern doesn't get
evaluated until runtime, as for example when it has a scalar reference
in it, like qr/$foo\N{NAME}/. We want the value for \N{NAME} that was
in effect at the point during the parsing phase that this regex was
encountered in, but we don't actually look at it until runtime, when
these bug reports show that it is gone. The solution is for the
tokenizer to parse \N{NAME}, but to compile it into an intermediate
value that won't ever be considered a metacharacter. We have chosen to
compile NAME to its equivalent code point value, and express it in the
already existing \N{U+...} form. This indicates to the regex compiler
that the original input was a named character and retains the value it
had at that point in the parse.
This means that \N{U+...} now always must imply Unicode semantics for
the string or pattern it appeared in. Previously there was an
inconsistency, where effectively \N{NAME} implied Unicode semantics, but
\N{U+...} did not necessarily. So now, any string or pattern that has
either of these forms is utf8 upgraded.
A complication is that a charnames handler can return a sequence of
multiple characters instead of just one. To deal with this case, the
tokenizer will generate a constant of the form \N{U+c1.c2.c2...}, where
c1 etc are the individual characters. Perhaps this will be made a
public interface someday, but I decided to not expose it externally as
far as possible for now in case we find reason to change it. It is
possible to defeat this by passing it in a single quoted string to the
regex compiler, so the documentation will be changed to discourage that.
A further complication is that \N can have an additional meaning: to
match a non-newline. This means that the two meanings have to be
disambiguated.
embed.fnc was changed to make public the function regcurly() in
regcomp.c so that it could be referred to in toke.c to see if the ... in
\N{...} is a legal quantifier like {2,}. This is used in the
disambiguation.
toke.c was changed to update some out-dated relevant comments.
It now parses \N in patterns. If it determines that it isn't a named
sequence, it passes it through unchanged. This happens when there is no
brace after the \N, or no closing brace, or if the braces enclose a
legal quantifier. Previously there has been essentially no restriction
on what can come between the braces so that a custom translator can
accept virtually anything. Now, legal quantifiers are assumed to mean
that the \N is a "match non-newline that quantity of times".
I removed the #ifdef'd out code that had been left in in case pack U
reverted to earlier behavior. I did this because it complicated things,
and because the change to pack U has been in long enough and shown that
it is correct so it's not likely to be reverted.
\N meaning a named character is handled differently depending on whether
this is a pattern or not. In all cases, the output will be upgraded to
utf8 because a named character implies Unicode semantics. If not a
pattern, the \N is parsed into a utf8 string, as before. Otherwise it
will be parsed into the intermediate \N{U+...} form. If the original
was already a valid \N{U+...} constant, it is passed through unchanged.
I now check that the sequence returned by the charnames handler is not
malformed, which was lacking before.
The code in regcomp.c which dealt with interfacing with the charnames
handler has been removed. All the values should be determined by the
time regcomp.c gets involved. The affected subroutine is necessarily
restructured.
An EXACT-type node is generated for the character sequence. Such a node
has a capacity of 255 bytes, and so it is possible to overflow it. This
wasn't checked for before, but now it is, and a warning issued and the
overflowing characters are discarded.
Jesse Vincent [Thu, 18 Feb 2010 19:03:23 +0000]
Update perldelta for 5.12 to warn vendors about deprecate.pm
(See [perl #72670])
Tony Cook [Thu, 18 Feb 2010 09:59:33 +0000]
rt #72866 - add magic to arrayrefs assigned to *Foo::ISA
The fix for rt #60220 (
26d68d86) updated the isa cache when an
arrayref was assigned to some *ISA, but didn't add the magic to the
new @ISA to catch any further updates to it. Add the magic, and
tests.
Nicholas Clark [Thu, 18 Feb 2010 15:38:11 +0000]
Avoid a panic from the UTF-8 length cache if the length overflows 32 bits.
Rather than storing a value, and having it wrap to a wrong value, treat such
lengths as "still unknown". This is a work around until a proper solution is
designed an implemented.
Nicholas Clark [Thu, 18 Feb 2010 15:23:49 +0000]
In Perl_reg_temp_copy(), ensure SvMAGIC() is NULL in the new REGEXP.
Change
c2123ae380a372d5 exposed the fact that Perl_reg_temp_copy() didn't
reset SvMAGIC() to NULL after block copying the "parent" regexp. The analagous
problem with SvSTASH() was fixed with change
b9ad13acb338e137.
Steve Hay [Thu, 18 Feb 2010 09:46:59 +0000]
Keep perl5115delta.pod up to date
Steve Hay [Thu, 18 Feb 2010 09:46:16 +0000]
Upgrade podlators from 2.3.0 to 2.3.1
Ensures that all $VERSIONs are bumped in files that have changed since
Perl 5.11.4.
Steve Hay [Thu, 18 Feb 2010 00:28:17 +0000]
Update perl5115delta.pod with today's changes
Steve Hay [Thu, 18 Feb 2010 00:17:57 +0000]
Upgrade CPAN from 1.94_55 to 1.94_56
Ensures that all $VERSIONs are bumped in files that have changed since
Perl 5.11.4.
Jesse Vincent [Wed, 17 Feb 2010 15:31:58 +0000]
Merge branch 'pod-perlrepository' of git://github.com/avar/perl into blead
* 'pod-perlrepository' of git://github.com/avar/perl:
Mention why it's a good idea to use topic branches for everything
Adjusted the git config user.name/user.email example for what 99% of users would like to do
Prefix commands universally with "%". Most of the POD used that convention but a few cases didn't
Don't recommend that people manually fiddle with their F<.git/config>, instead they should edit it with L<git-config(1)>
Note that checking out from git via http:// is at least 4x slower than git://
Jesse Vincent [Wed, 17 Feb 2010 14:45:30 +0000]
Make the new Socket implementation of inet_pton consistent with the existing Socket6 implementation of inet_pton.
Fix for release-blocking ticket [perl #72884]
> |https://rt.cpan.org/Ticket/Display.html?id=52497|
4411113f|NJH/IO-Socket-Multicast6-0.03.tar.gz |
I'll describe what's happening here, and leave it to everyone else to decide
which of the interacting events is the bug, or where the fix might be.
So the *entire* change is this:
commit
4411113f31b3f00171bb335092b02104d29d7cd7
Author: Rafael Garcia-Suarez <rgarciasuarez@gmail.com>
Date: Fri Mar 27 13:19:16 2009 +0100
Add inet_pton and inet_ntop to the list of functions exported by Socket
diff --git a/ext/Socket/Socket.pm b/ext/Socket/Socket.pm
index 6b268ef..7d130ba 100644
--- a/ext/Socket/Socket.pm
+++ b/ext/Socket/Socket.pm
@@ -198,6 +198,7 @@ use XSLoader ();
@ISA = qw(Exporter);
@EXPORT = qw(
inet_aton inet_ntoa
+ inet_pton inet_ntop
sockaddr_family
pack_sockaddr_in unpack_sockaddr_in
pack_sockaddr_un unpack_sockaddr_un
and test fails like this:
$ /home/nick/Sandpit/snap5.9.x-v5.11.4-85-g95c9bfa/bin/perl5.11.4 -Mblib t/35mcastsend.t
1..11
Constant subroutine IO::Socket::Multicast6::AF_INET6 redefined at /home/nick/Sandpit/snap5.9.x-v5.11.4-85-g95c9bfa/lib/perl5/5.11.4/Exporter.pm line 64.
at /home/nick/.cpan/build/IO-Socket-Multicast6-0.03-CqGAYT/blib/lib/IO/Socket/Multicast6.pm line 10
Prototype mismatch: sub IO::Socket::Multicast6::AF_INET6 () vs none at /home/nick/Sandpit/snap5.9.x-v5.11.4-85-g95c9bfa/lib/perl5/5.11.4/Exporter.pm line 64.
at /home/nick/.cpan/build/IO-Socket-Multicast6-0.03-CqGAYT/blib/lib/IO/Socket/Multicast6.pm line 10
Constant subroutine IO::Socket::Multicast6::PF_INET6 redefined at /home/nick/Sandpit/snap5.9.x-v5.11.4-85-g95c9bfa/lib/perl5/5.11.4/Exporter.pm line 64.
at /home/nick/.cpan/build/IO-Socket-Multicast6-0.03-CqGAYT/blib/lib/IO/Socket/Multicast6.pm line 10
Prototype mismatch: sub IO::Socket::Multicast6::PF_INET6 () vs none at /home/nick/Sandpit/snap5.9.x-v5.11.4-85-g95c9bfa/lib/perl5/5.11.4/Exporter.pm line 64.
at /home/nick/.cpan/build/IO-Socket-Multicast6-0.03-CqGAYT/blib/lib/IO/Socket/Multicast6.pm line 10
ok 1 - use IO::Socket::Multicast6;
ok 2 - Create IPv4 multicast socket
ok 3 - Combined IPv4 destination address and port
ok 4 - Separate IPv4 destination address and port
Bad arg length for Socket::pack_sockaddr_in, length is 16, should be 4 at t/35mcastsend.t line 26.
The warnings aren't actually really relevant. The "problem"s are:
IO::Socket::Multicast6 isa IO::Socket::INET6
IO::Socket::INET6 isa IO::Socket
so they chose to inherit all behaviour from IO::Socket. In turn
IO::Socket isa IO::Handle isa Exporter.
IO::Socket chooses to export everything that Socket does:
sub import {
my $pkg = shift;
if (@_ && $_[0] eq 'sockatmark') { # not very extensible but for now, fast
Exporter::export_to_level('IO::Socket', 1, $pkg, 'sockatmark');
} else {
my $callpkg = caller;
Exporter::export 'Socket', $callpkg, @_;
}
}
So, this means that all those choices and delegation of behaviour (and
responsibility) means that all those modules export whatever Socket exports.
OO modules.
So, now they also export inet_pton and inet_ntop.
The test is careful to only import what it needs:
use strict;
use Socket6 qw/ inet_pton pack_sockaddr_in6/;
use Socket qw/ pack_sockaddr_in /;
use Test::More tests => 11;
in particular, it wants inet_pton from Socket6, and only pack_sockaddr_in
from Socket. So at that point, main::inet_pton is Socket6::inet_pton
Then it does this, correctly in a BEGIN block:
BEGIN { use_ok( 'IO::Socket::Multicast6' ); }
The side effect of this is to import all exports from IO::Socket::Multicast6.
Which, from the above chain of inheritance, is @Socket::EXPORT.
So at this point, main::inet_pton is rebound to Socket::inet_pton
And the test fails, because they differ, and it expected (and wanted)
Socket6::inet_pton(). Socket6::inet_pton() returns 4 bytes for AF_INET, 16
bytes for AF_INET6. Socket::inet_pton() returns 16 for both.
Socket::pack_sockaddr_in() wants 4 bytes.
Now, to add to the fun:
use IO::Socket::Multicast6 ();
and
use IO::Socket::Multicast6;
of course mean different things. The former suppresses all exports.
It turns out that with Test::More::isa_ok() has no way of doing the former:
$ cat use_ok.pl
use warnings;
use strict;
use Test::More tests => 3;
package clash;
BEGIN { main::use_ok 'Socket' };
package crunch;
BEGIN { main::use_ok 'Socket', () };
package biff;
BEGIN { main::use_ok 'Socket', 'sockaddr_family' };
package main;
sub dump_lowercase_keys {
my $package = shift;
print "For package $package:\n";
no strict 'refs';
print " $_\n" foreach sort grep {!tr/A-Z//} keys %{"${package}::"};
print "\n";
}
dump_lowercase_keys $_ foreach qw (clash crunch biff);
__END__
$ perl use_ok.pl
1..3
ok 1 - use Socket;
ok 2 - use Socket;
ok 3 - use Socket;
For package clash:
inet_aton
inet_ntoa
pack_sockaddr_in
pack_sockaddr_un
sockaddr_family
sockaddr_in
sockaddr_un
unpack_sockaddr_in
unpack_sockaddr_un
For package crunch:
inet_aton
inet_ntoa
pack_sockaddr_in
pack_sockaddr_un
sockaddr_family
sockaddr_in
sockaddr_un
unpack_sockaddr_in
unpack_sockaddr_un
For package biff:
sockaddr_family
So, there's no clean way to rewrite that test with use_ok to suppress imports.
So, to summarise, it's due to
A cascade of modules blindly exporting everything that Socket exports
Socket::inet_pton() and Socket6::inet_pton() differing in behaviour
A test that fails to realise that it's importing everything via use_ok
This one ranks as blocker because:
Socket::inet_pton() and Socket::inet_ntop() are not in any stable release
Hence we have the option to change them if we do it *NOW*.
I think that the right fix is the appended patch. This makes the new Socket
implementation of inet_pton consistent with the existing Socket6
implementation of inet_pton.
Offer Kaye [Wed, 17 Feb 2010 14:37:10 +0000]
Documentation fix in perldoc's usage statement. (-A -> -v)
Reported in: <
5694251002150108m12a3d68flb196f14a45103149@mail.gmail.com>
Steve Hay [Wed, 17 Feb 2010 09:27:51 +0000]
Change my email address from work to home
Steve Hay [Tue, 16 Feb 2010 23:23:55 +0000]
Keep perl5115delta.pod up to date
Ævar Arnfjörð Bjarmason [Tue, 16 Feb 2010 18:19:18 +0000]
Mention why it's a good idea to use topic branches for everything
Most of this is derived from Dan Golden's E-Mail to me on
perl5-porters with the subject "[PATCH] Add comments to gv.c about
variable implementation" where I'd submitted a patch without using a
topic branch.
Ævar Arnfjörð Bjarmason [Tue, 16 Feb 2010 17:46:46 +0000]
Adjusted the git config user.name/user.email example for what 99% of users would like to do
Almost nobody wants to set user.name/user.email on a per-repository
basis as the previous example suggested. I put in a new example that
shows how to use C<git config --global> along with an example of how
to set user.email just for F<perl-git>
I replaced Leon Brocard's name with mine because he didn't fit into
the multiple E-Mail address example and I couldn't think of another
example.
Ævar Arnfjörð Bjarmason [Tue, 16 Feb 2010 17:36:58 +0000]
Prefix commands universally with "%". Most of the POD used that convention but a few cases didn't
Ævar Arnfjörð Bjarmason [Tue, 16 Feb 2010 17:33:43 +0000]
Don't recommend that people manually fiddle with their F<.git/config>, instead they should edit it with L<git-config(1)>
Jesse Vincent [Tue, 16 Feb 2010 17:28:01 +0000]
Fix another place where I typoed book's name
Jesse Vincent [Tue, 16 Feb 2010 17:26:30 +0000]
Doc fix. Thanks to Bram.
Ævar Arnfjörð Bjarmason [Tue, 16 Feb 2010 17:23:25 +0000]
Note that checking out from git via is at least 4x slower than git://
Tested on a 10 MiB/s connection in London, although this is also
IO/CPU bound since Git needs to resolve deltas and check out after the
repository has made it across the wire.
Nicholas Clark [Tue, 16 Feb 2010 16:16:33 +0000]
Fix #72850 - reading $! shouldn't SEGV if Strerror(errno) returns NULL.
This can happen on some OSes for out of range errno values. The bug was
introduced with
0097b436152452e4, which in turn fixed #61976.
Test case by Steve Peters.
Philippe Bruhat (BooK) [Tue, 16 Feb 2010 16:16:47 +0000]
$name =~ /Phill?ipp?e?/ && $name eq "Philippe" # FAIL